发表机构
Beijing University of Posts and Telecommunications; Beijing Key Laboratory of Multimodal Data Intelligent Perception and Governance（北京邮电大学; 北京多模态数据智能感知与治理重点实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对细粒度识别任务中层次梯度冲突致训练不稳定问题，提出无参数的FlexiGrad方法，通过调节反向传播梯度交互，去除有害冲突成分，强化共享方向，集成现有架构，提高多粒度精度。
AI 中文摘要
许多细粒度识别任务包含层次标签，如目、科和物种。虽然这种监督应有益，但联合优化所有级别通常会导致训练不稳定，因为粗粒度和细粒度分类器对共享主干施加不一致的梯度。这种层次梯度冲突会阻止模型学习连贯的从粗到细的表示。本文提出FlexiGrad，一种简单且无参数的方法，在反向传播期间调节梯度交互。FlexiGrad在任务不一致时仅去除有害冲突成分，在部分一致时通过平滑的层次感知加权函数强化共享方向。这产生稳定优化并保留全局结构和细粒度判别线索。FlexiGrad无需修改即可集成到现有架构中，同时提高了在CUB - 200 - 2011、FGVC - 飞机和斯坦福汽车数据集上的多粒度精度。代码将在PRIS - CV/FlexiGrad上可用。
英文摘要
Many fine-grained recognition tasks contain hierarchical labels such as order, family and species. Although this supervision should be beneficial, jointly optimising all levels often leads to unstable training because coarse and fine classifiers impose inconsistent gradients on the shared backbone. This hierarchical gradient conflict prevents the model from learning a coherent coarse-to-fine representation. In this paper, we propose FlexiGrad, a simple and parameter-free method that regulates gradient interactions during backpropagation. FlexiGrad removes only the harmful conflicting component when tasks disagree and reinforces the shared direction when they partially agree through a smooth hierarchy-aware weighting function. This produces stable optimisation and preserves both global structure and fine-grained discriminative cues. FlexiGrad integrates into existing architectures without modification while improves multi-granularity accuracy on CUB-200-2011, FGVC-Aircraft and Stanford Cars. The code will be available at PRIS-CV/FlexiGrad.
